serve-gr00t-policy
Starts, reuses, verifies, and explicitly stops the local GR00T inference server used by Isaac Lab-Arena Gr00tRemoteClosedloopPolicy experiments. Use when…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-06 00:14
Agents & models
25,709 results
setup-agent-workflow
Scaffold a repo for agent-driven workflows: creates AGENTS.md and CLAUDE.md with the import-pointer pattern, and records where issues/intents live (GitHub,…
skill-create
Audits an Agent Skill directory against Cursor skill authoring conventions (frontmatter, description, structure, progressive disclosure, scripts,…
open-second-brain
Use Open Second Brain to read, write, and maintain an agent-owned second brain in an Obsidian-compatible Markdown vault. The agent owns one top-level directory…
find-skills
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express…
p-clipboard-copy
Pull copy-worthy fragments out of the ongoing session — commands, code blocks, file paths, URLs, IDs, queries, message drafts, error snippets, diffs — and push…
find-skills
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express…
opencode-agents
Define and configure OpenCode agents with proper permissions, tools, and behaviors. Use for agent creation, tool access control, permission patterns, and mode…
doc-to-skill
Use when the primary outcome is creating a new portable Agent Skill directory from one local documentation file, documentation directory, or Git repository,…
output-style-creator
Create, validate, and diagnose Claude Code output styles — the file that sets Claude's role, tone, and response format for every turn. Use when asked to write,…
parallel-work
Shapes for running many sub-agents at once — fan-out with a barrier, maker/checker, generate-and-filter, tournaments for ranking, hypothesis fan-out for root…
huxishu
Hu Xishu (1898-1984), professor at Beijing College of Traditional Chinese Medicine, classical formula clinician and educator. Core mental models: the six channels derive from the eight principles; formula-pattern is the tip of pattern differentiation; the struggle between upright qi and pathogenic qi; stomach qi as the foundation. Decision heuristics: fever with aversion to cold to differentiate yin and yang; sweating indicates Gui Zhi Tang, no sweating indicates Ma Huang Tang; when even one sign is present, it is a Chai Hu pattern; the three stages of Cheng Qi Tang; the method of exclusion to differentiate half-exterior half-interior. Trigger words: "Hu Xishu", "Old Hu", "classical formula Hu Xishu", "six channels and eight principles", "how does Hu Xishu see it", "formula-pattern correspondence", "Shanghan Lun Hu Xishu"…
session-recall
Search past Claude sessions for context. Use when asked "where did we leave off?", "what did we discuss about X?", "find a previous conversation", or when…
building-anime-prompts
Builds structured anime and manga prompts for Midjourney Niji mode, matching artist styles, SREF codes, and technical parameters to user requests. Activates…
developing-agents
Guides development of Claude Code plugin agents, including agent file structure, YAML frontmatter fields, system prompt design, triggering conditions, and…
integrating-mcp
Guides integration of Model Context Protocol (MCP) servers into Claude Code plugins, including server types (stdio, SSE, HTTP, WebSocket), .mcp.json…
character
Give your Claude Code employee a real personality. character, tone of voice, humor, and values. Runs a short research-backed interview, then writes a dedicated…
orchestrating-fusion
Coordinate a read-only lead and implementation executor when role separation is requested.
policy
Analyze and author MCP tool-call policy rules without hand-editing access.json. Use when adding, linting, or removing auto_approve/deny/require_approval rules…
dolt-mcp-vcs
Universal Dolt version-control workflow. Step 0 auto-detects what kind of Dolt stack is present — every make and model (classic Dolt repo, live dolt sql-server…
outreach-connectors
Inspect the bundled Intent Outreach connector registry without running research or enrichment. Use when a user asks which providers are configured or wants a…
j-rig-eval
Evaluate an Agent Skill with JRig and return an evidence-backed ship, warn, block, or obsolete-review decision. Use when grading, scoring, release-gating,…
autoresearch
Generate deterministic prototype hypotheses, simulated experiment findings, and aggregate confidence summaries. Use when exercising an offline research-loop…
gui_agent
Simulate element detection, UI actions, and form-validation results without opening a browser or changing a page. Use when exercising the GUI Agent contract…